Setting threshold and baseline in qPCR, and why your Cq depends on it
Cq looks like a measurement. It is really the result of two settings, the baseline and the threshold, and the software's defaults are not always right for your assay.
Cq (quantification cycle) is the cycle at which a well's baseline-corrected fluorescence crosses the threshold. The baseline is the range of early cycles, before amplification is detectable, used to estimate and subtract background; the threshold is a fluorescence level set within the exponential phase, where curves on a log plot are straight and parallel. Set the baseline to end a couple of cycles before the earliest curve rises, place the threshold in the parallel log-linear region above noise, and keep the same threshold for one target across every plate you compare.
The baseline: removing background
Every well has some background fluorescence, and it drifts. The baseline is a window of early cycles, typically starting around cycle 3, used to fit and subtract that background. If the window extends into cycles where amplification has started, the software subtracts real signal, curves flatten and Cq shifts late. For high-abundance targets that rise early, shorten the baseline window.
The threshold: where Cq is read
View the curves on a log scale. In the exponential phase they become straight lines that should run parallel if efficiencies match. Place the threshold in that region, clear of baseline noise and well below the plateau. Because the curves are parallel there, the differences between Cq values, which is all ΔΔCt uses, stay the same wherever in the region you place it.
Common mistakes
- Different thresholds per plate for the same target, then comparing Cq across plates.
- Automatic baselines on early-rising wells, which subtract signal.
- A threshold in the noise, giving erratic Cq for NTCs and late wells.
- Ignoring curve shape: a flat or odd-shaped curve with a Cq is still a failed reaction.
Doing it interactively
In CycleView, open the run file, switch to the log view, and drag the threshold line for each target. The baseline window is set per target too. Cq values, replicate SDs and downstream ΔΔCt update as you drag, so you can see exactly how sensitive your result is to the setting, and report the values you used in your methods.
Open your run on your own laptop
CycleView opens QuantStudio, LightCycler, Rotor-Gene and RDML files on Mac and Windows. The viewer is free, and nothing is uploaded.
Get CycleView
Mac and Windows · runs locally · for research use, not for diagnosis.
FAQ
Ct, Cp or Cq: which should I report?
The MIQE guidelines recommend Cq as the vendor-neutral term. State the method used to determine it.
Should I use the software's automatic threshold?
It is a starting point. Check it on the log view and fix one threshold per target across plates you compare.
How much replicate variation is acceptable?
Technical replicates within about 0.3 cycles of each other are commonly considered good; larger spreads deserve a look at the curves.